a review of collaboration of multi-cloud – an effective use of cloud computing

7/30/2019 A Review of Collaboration of Multi-Cloud An Effective Use of Cloud Computing

1/4

International Journal of Application or Innovation in Engineering& Management (IJAIEM)Web Site: www.ijaiem.org Email: [email protected], [email protected]

Volume 2, Issue 3, March 2013 ISSN 2319 - 4847

Volume 2, Issue 3, March 2013 Page 328

ABSTRACTCloud computing is an emerging trend which can transform information technology and the way it is used. Cloud service

provider basically provides Software, Platform and infrastructure as service to the user on pay-per-use basis. . Cloud nowadays

is gaining huge attention due to its characteristics as it can be modeled as per the requirement of the company or the user. The

user when gains a service from a cloud gets vendor lock-in and has to avail all services from a single cloud provider. Asorganizations are now adopting the cloud environment the cloud service providers are now moving towards a new concept i.e.

multi cloud where a user can utilize services from multiple cloud service provider. These multi-clouds are organized by aagreement between different service provider to provide a low cost functionalities to the client. This paper will review the work

done by the researcher in the field of development of architecture, standardized protocol, the interoperability of services

provided such as software, platform and infrastructure and the security issue which must be evaluated while shifting from theparadigm of cloud computing to multi-cloud.

Keyword: Cloud service Provider, Proxy based framework, SPL, Virtual machine

1.INTRODUCTIONThe term cloud computing can be defined as a system that is concerned with the integration, virtualization,standardization, and management of services and resources. The benefits of cloud computing include minimized

capital expenditure, utilization and efficiency improvement, high computing power, location and device independenceand finally very high scalability [1]. Cloud computing brings a scenario in the field of IT that gives a model where auser who wants to gain access to the software without licensing it, platform to run this software and the infrastructurecan access these services on pay-per-use basis. The cloud platform also provides a large amount of data storage to theuser who can utilize it. Moving data into the cloud offers great convenience to users since they dont have to care aboutthe complexities of direct hardware management. The pioneer of cloud computing vendors, Amazon Simple StorageService (S3), and Amazon Elastic Compute Cloud (EC2) are both well-known examples [2]. Different approaches havebeen investigated that encourage the owner to outsource the data, and offer some sort of guarantee related to theconfidentiality, integrity, and access control of the outsourced data[3]. The user who gain access to the cloud servicegain all these services but the user gets vendor lock-in and has to use all the service by this particular cloud serviceprovider if users want to gain access to another cloud service provider for more effective and low cost management userhas to authenticate to a particular service provider in this way user has to use multi-service provider on individual basisand pay separately for the service to each provider. The scenario of multi-cloud presents a model called collaboration ofmulti-cloud where the user vendor lock-in can be abolished with an agreement between the various cloud serviceprovider that an authorized user of a particular cloud service provider can gain access to different service provider asper his requirement and cost management. To avoid the vendor lock-in syndrome, SaaS must be portable on top ofvarious cloud PaaS and IaaS providers. This portability allows the migration from one provider to another in order totake advantage of cheaper prices or better qualities of services (QoS) [4]. Examples of cloud mashup centre are IBMsMashup Center, Appirio Cloud Storage and Force.com for the Google App engine. The concerns over the establishmentof such collaboration are that the architecture, protocols and other platform are on the research level. Another aspect isthat it might be difficult that the various cloud service provider can get into collaboration so that a user can gain accessto different service provider while he/she is an authenticated user of a single cloud service provider. In a multi-providerhosting scenario, the Service Provider is responsible for the multi-cloud provisioning of the services. Thus, the Serviceprovider contacts the possible Infrastructural Providers, negotiates terms of use, deploys services, monitors theiroperation, and potentially migrates services (or parts thereof) from misbehaving Infrastructural Providers.

Infrastructural Providers are managed independently and placement on different providers is treated as multipleinstances of deployment [5]. However, Apart from this issue the main role has to been played by the researchers todevelop a mechanism to bring this collaboration or mashup centre into a real world for the standardized and costeffective of use cloud computing. The issue of security will also get generated as soon these mashup centre starts

A Review of Collaboration of Multi-Cloud AnEffective Use of Cloud Computing

Swaraj P. Thakre1, Prof. Nitin R. Chopde2

1M.E.Scholar, 2M.E. (Computer Engineering)1,2Department of Computer Science and Engineering,

1,2G.H. Raisoni College of Engineering and Management, Amravati.


2/4




working which also must be look around as the service provider should not get it as a threat while implementing thesecentre. This paper will present a review of all these aspects which have been researched by the researcher community tomake this collaboration work and a real world scenario can be presented.1.1 LITERATURE REVIEW

This is a review paper based on the research work done by the researcher in the field of a new environment in cloud

computing i.e. the collaboration of multi-cloud. This will give an overview of the techniques which will be helpful forshifting from the single cloud architecture to multi-cloud architecture, a security model and cost effectiveness of multi-cloud compared to a cloud. Multi-Cloud computing has many advantages such as it provides usage of data from variousclouds, the ability of choice for the user, stops vendor lock-in and synchronization between different cloud serviceproviders with cost optimization. The main issue in implementing multi-cloud is its working in a distributedenvironment as the services are to be collaborated with different cloud service providers to make it possible aframework is laid in the research work of Collaboration Framework for Multi-cloud Systems [6] which specify theuse of proxies at different level of collaboration. These proxies can be implemented by the cloud service provider or canbe set by the institutions\organization so as to gain service from collaborated service providers. These proxies can also bused to have a secure communication between the client and the service provider.To protect data at rest and data intransit, proxies must provide a trusted computing platform that prevents malicious software from taking control andcompromising sensitive client and cloud application data [6]. This also deals with the security aspect of the cloudcomputing. The cloud services have been classified as software as a service (SaaS), Platform as a service (PaaS) andInfrastructure as a Service (IaaS) it becomes important that the cloud service providers must be able to provide theseservices on distributed environment of multi-cloud for that purpose research work of A Federated Multi-Cloud PaaSInfrastructure [4] can be effective as it provides a platform for various services to be provides in a collaboratedparadigm. It is also important that the cost effectiveness of multi-cloud must be considered before shifting towards anew paradigm to solve this issue research work of Cloud Brokering Algorithm [10] has given an algorithm based onthe Virtual infrastructure in cloud environment which will effectively determine the allocation of VM both on static anddynamic basis. This paper is based on review of the technique that will proof to be efficient while shifting towards themulti-cloud environment. All the factor included in this paper are the research work done in the fields which are themajor concern whenever a new technology is to be implemented it includes the framework, platform for newtechnology to be implemented and the cost effectiveness on the side of the consumer.

2.METHODOLOGYThis section covers the proposed system by the researchers which have been discussed in the literature review.

2.1Proxy Based FrameworkA proposed proxy-based multi-cloud computing framework allows dynamic, on the-fly collaborations and resource

sharing among cloud-based services, addressing trust, policy, and privacy issues without pre-established collaborationagreements or standardized interfaces [6].It includes the use of proxy in multi-cloud environment in various form theseare

2.1.1Cloud-hosted proxyIn this scenario the cloud service provider host proxies within its infrastructure administer and manage the proxies

and will handle the service request from the client who wants to access these proxies.2.1.2Proxy as a service

Here the proxy is been deployed as an autonomous cloud. Multiple cloud service providers with collaboration canmanage this proxy or a third party proxy service provider can manage it for the cloud service providers.

2.1.3Peer-to-peer proxyProxy can also be interacted on peer-to-peer network which is managed by the proxy service provider or cloud

service provider those who have an agreement of collaboration.2.1.4On-premise proxy

The client himself can host proxies within infrastructural domain and manage it in administrative domain. The userwho wishes to use proxies will have to deploy it on premise proxies and the service providers that wish to collaboratewith other service provider will have to implement it within the service-requesting client domain.

2.2Security IssuesSharing applications that process critical information with different tenants without sufficient proven security

isolation, security SLAs or tenant control, results in "loss-of-control" and lack of trust problem[7].Using proxiesmoves the trust boundary one step further: clients and CSPs now must establish trust relationships with proxies, whichincludes accepting a proxys security, reliability, availability, and business continuity guarantees [6]. A trustworthycollaboration must be set between the client and Cloud service provider which will help in management and

administering proper communication. In this framework different types of proxies network is been explained some areCSPs side and some are established on client side this states the control over the assets while processing proxies andsimilarly using proxies that are within the domain of cloud service provider exercise its control over proxiesadministration. Proxy network is a potential platform for developing proxy based security architecture. Data


3/4




confidentiality on transmission in proxy based network can be achieved using Transport Layer Security Protocol. Someother technologies that can be used to provide security are warrant-based proxy signature for delegation signing rightsto provide authentication to the proxies and simple public-key infrastructure can provide secure access andauthentication.

2.3A Federated Multi-Cloud PaaS InfrastructureThis federated infrastructure offers some solutions to the problems such as portability, interoperability and geo-diversity for management of both SaaS and PaaS. The different layers of a cloud environment (IaaS, PaaS, and SaaS)

provide dedicated services. Although their granularity and complexity vary, we believe that a principled definition ofthese services is needed to promote the interoperability and federation between heterogeneous cloud environments [4].

This federated infrastructure is based on following three models:2.3.1Open Service Model

The different layers of a cloud environment (IaaS, PaaS and SaaS) provide dedicated services. Although theirgranularity and complexity vary, we believe that a principled definition of these services is needed to promote theinteroperability and federation between heterogeneous cloud environments [4]. A Service Component Architecture isdesigned for running service-oriented distributed applications. Its supports interaction between different protocols forthis it has a notion of binding. Hence SCA is used for both the definition of services in federated PaaS and services ofSaaS.

2.3.2Configurable Federated Multi-PaaS InfrastructureThis federated multi-PaaS architecture depends on configurable kernel which can be implemented in concrete cloud

environment. A Software product line can be defined as a set of software-intensive system that share a common,managed set of features and that are developed from a common set of core assets in a prescribed way [4]. The basic ideaof defining the software product line is to capture the points of variability between the cloud environments andimplement this SPL as a component of SCA.

2.3.3Infrastructure ServicesA generic architecture has been laid down by the definition of Service component architecture and configurable

Kernel in this environment a cloud that hosts SaaS is considered as a node and configurable Kernel as an instance forparticular cloud. The service list first allocates the resources on all nodes and then deploy the configurable kernel andapplications on each node the second step involves the deployment of instances of configurable kernel and applicationson particular node as both the PaaS and SaaS are based on service component architecture they can be deployed eitheron the kernel level or on the application level.

2.4Cloud Brokering Algorithm

FIGURE 1: Cloud brokering architecture overview [10]

In cloud computing, subscribers have to pay the service providers for the storage service. This service does not onlyprovides flexibility and scalability for the data storage, it also provide customers with the benefit of paying only for theamount of data they need to store for a particular period of time, without any concerns for efficient storage mechanismsand maintainability issues with large amounts of data storage [8]. The cost effectiveness of deployment of cloud


4/4




depends upon the deployment of virtual infrastructure it also affects whether it is static or dynamic deployment. Manyresearchers focus only on static deployment where the user of service providers condition does not change but in somecases the deployment has to be changed according to the time factor so as to be cost effective. Cloud computing can beclassified as a new paradigm for the dynamic provisioning of computing services supported by state-of-the-art datacenters that usually employ Virtual Machine (VM) technologies for consolidation and environment isolation purposes

[9]. The optimal deployment of VM is an important factor for cost effectiveness of cloud service provider. Thechallenge is for determining the provisioning of virtual infrastructure as it should not be over\under provision. Thesystem architecture given in fig. 1 gives an improved model of dynamic scheduler of multi-cloud brokering algorithm.

This broker consist of service description, cloud broker and cloud service provider. The user can request the servicedescriptor template for virtual infrastructure which consists of number of VM to be deployed among available cloud.

The cloud broker which is an intermediate between service descriptor and cloud service provider has to perform twomajor tasks i.e. placement of virtual resources and management of these resources. The scheduler is responsible for theallotment of virtual infrastructure in available clouds. This situation is been implemented in static and dynamicenvironment. In the static approach, the placement decision is made once, according to the current user and pricingconditions. The dynamic approach is suitable for variable conditions (e.g., variable resource prices, required virtualresources, or cloud provider resources availability), so a new placement decision can be made when conditions change[10].

3.CONCLUSIONThis paper reviewed all those technique that are area of concern when a paradigm is to be changed the framework orthe architecture to built the environment, the platform on which the services are to be shared and at last the marketpoint of view that is its cost effectiveness compared to the available. The multi-cloud environment can end the vendorlock-in of the consumer which is attained in the single cloud. The major area of concern in this field is the agreementbetween the cloud service providers for collaboration of their services in multi-cloud. The consumer will get highlybenefited with multi-cloud environment and obtain service based on his preferences and requirement and not based onhis cloud service provider.

REFERENCES[1] R. Thandeeswaran, S. Subhashini, N. Jeyanthi1, M. A. Saleem Durai, Secured Multi-Cloud Virtual Infrastructure

with Improved Performance, cybernetics and information technologies XII, ( 2), pp. 11-22, 2012[2] Cong Wang, Student Member, Qian Wang, Student Member, Kui Ren, Senior Member, Ning Cao, and Wenjing

Lou, Toward Secure and Dependable Storage Services in Cloud Computing, IEEE transactions on servicescomputing, V, (2), 2012.

[3] Ayad Barsoum and Anwar Hasan, Enabling Dynamic Data and Indirect Mutual Trust for Cloud ComputingStorage Systems, IEEE transactions on parallel and distributed systems.

[4] Fawaz Paraiso, Nicolas Haderer, Philippe Merle, Romain Rouvoy, Lionel Seinturier, A Federated Multi-CloudPaaS Infrastructure, 5th IEEE International Conference on Cloud Computing pp.392 399, 2012

[5] Ana Juan Ferrer, Francisco Hernndezb, Johan Tordsson , Erik Elmroth, Ahmed Ali-Eldin, Csilla Zsigri, RalSirvent, Jordi Guitart, Rosa M. Badia, Karim Djemamee, Wolfgang Ziegler, Theo Dimitrakos, Srijith K. Nair,George Kousiouris, Kleopatra Konstanteli, Theodora Varvarigou, Benoit Hudzia, Alexander Kipp, Stefan Wesnerj,Marcelo Corrales, Nikolaus Forg, Tabassum Sharif, Craig Sheridan, OPTIMIS: A holistic approach to cloudservice provisioning, Future Generation Computer Systems ELSEVIER pp. 6677, 2012.

[6] Mukesh Singhal and Santosh Chandrasekhar, Tingjian Ge, Ravi Sandhu and Ram Krishnan, Gail-Joon Ahn, andElisa Bertino Collaboration in Multicloud Computing Environments: Framework and Security Issues, Publishedby the IEEE Computer Society IEEE, 2013.

[7] Mohamed Almorsy, John Grundy, and Amani S. Ibrahim, TOSSMA: A Tenant-Oriented SaaS SecurityManagement Architecture, 5th IEEE Conference on Cloud computing IEEE, 2012.

[8] Yashaswi Singh, Farah Kandah, Weiyi Zhang, A Secured Cost-effective Multi-Cloud Storage in CloudComputing, IEEE INFOCOM Workshop on Cloud Computing, 2011.

[9] Anton Beloglazov, Jemal Abawajy, Rajkumar Buyyaa, Energy-aware resource allocation heuristics for efficientmanagement of data centers for Cloud computing, Future Generation Computer Systems ELSEVIER , pp. 755768, 2011.

[10] Jose Luis Lucas-Simarro, Rafael Moreno-Vozmediano, Ruben S. Montero and Ignacio M. Llorent, Costoptimization of virtual infrastructures in dynamic multi-cloud scenarios, Concurrency and Computation: practiceand experience Concurrency Computat.: Pract. Exper. Published online in Wiley Online Library(wileyonlinelibrary.com). 2012.

a review of collaboration of multi-cloud – an effective use of cloud computing

Documents